APPARA TUS FOR SCIENTIFIC DEMONS TRA TION 165 



the right nearly as far as it will go ; if not, each full revolution 

 to the left will lower the potential from \\ to 2 volts. The 

 arc having been struck, the lamp is run a little till it has 

 attained the length desired; when the regulator is turned 

 very gently to the left until the mechanism gives a * feed,' 

 with the accompanying click. When this is adjusted, the 

 lamp will regulate itself. Only if the current is very small 

 for the lamp, may it be needful to push down the lower carbon 

 a very little, so that the arc struck may be from the first 

 shorter : afterwards the procedure will be the same. For 

 smaller currents smaller carbons will of course be used. Hard 

 carbons should be used for the negative, and soft-cored ones 

 for the positive pole. The collars against which the lower 

 carbon is forced by the spring are provided of various sizes 

 according to the carbons, so that the point protrudes the 

 required distance. 



The only purpose for which this arrangement is not suit- 

 able, is vapourising substances for spectrum projections, which 

 require vertical carbons and the crater at the bottom. The 

 needful modification which allows the Brockie lamp to be 

 used for this also, is described in the proper place. 



The Brockie lamp does, however, require a continuous 

 current ; and I confess that I dread the growing tendency to 

 alternate currents, as regards all work in public demonstra- 

 tion. A lamp can easily be made to work with such currents ; 

 but as there is no positive * crater,' the special radiation from 

 one cannot be had, and we can only hide one of the two 

 luminous poles at the expense of fully half the light. This 

 is a real difficulty, only at present to be overcome by storage 

 batteries. Moreover the manipulation occasionally necessary 

 in all optical work, is liable with alternate currents to produce 

 very unpleasant consequences. Should a continuous current 

 of, say, 10 amperes at 50 volts give an accidental shock, though 

 unpleasant to most persons at the moment, it would have no 

 further effect ; but the same current taken alternately would 



